DESCRIPTION (Verbatim from the Applicant's Abstract): Two important changes regarding contrast enhanced MRI (ceMRI) of the heart have recently occurred. First, it is now recognized that segmented k-space inversion-recovery imaging pulse sequences result in image intensities in "hyperenhanced" regions which are typically 500 percent higher than in "non hyperenhanced" regions, greatly reducing observer subjectivity. Second, recent data suggest that healed myocardial infarcts hyperenhance. In preliminary studies, the applicants found that ceMRI detects both acute and chronic infarcts with a sensitivity approaching that of serum assays for cardiac enzymes. Unlike cardiac enzymes which are cleared from the blood in a few days, however, ceMRI provides a permanent record of infarction, localizes the infarct to a specific coronary artery territory, and can be combined with cine MRI to document the effect of the infarct on wall motion with perfect registration. In addition, preliminary results suggest that hyperenhancement may be specific for ischemic injury. The applicant studied patients with non-ischemic cardiomyopathy determined by coronary angiography. Despite profound ventricular dysfunction, hyperenhancement was not observed in these patients. Detailed analysis of the data from patients with ischemic and non-ischemic cardiomyopathy strongly suggested that, for patients with ischemic disease, the presence or absence of irreversible injury. To investigate this further, the applicants examined wall motion in patients before and after revascularization by CABG or PTCA. For the 804 segments with a baseline wall motion abnormality, the likelihood of recovery of wall motion after revascularization was strongly predicted by the presence or absence of hyperenhancement. These preliminary data underscore that ceMRI in combination with cine MRI can provide detailed diagnostic information. The applicants propose to establish the sensitivity of ceMRI to detect chronic infarction in patients (Aim 1), to test the hypothesis that recovery of wall motion following revascularization is predicted by ceMRI (Aim 2), and determine if ceMRI can distinguish between patients with ischemic and non-ischemic cardiomyopathy (Aim3).